Background: Divarasib, a covalent inhibitor targeting the Kirsten rat sarcoma virus oncogene homologue glycine-to-cysteine mutation at position 12 (KRAS G12C), is currently in clinical development for Non Small Cell Lung Cancer (NSCLC) treatment, with various combination partners, such as the Src homology region 2 domain-containing phosphatase-2 (SHP2) inhibitor migoprotafib. A quantitative systems pharmacology (QSP) model is essential to quantitatively assess the single-agent and combination pharmacodynamic (PD) effects observed in the preclinical setting to reveal the potential mechanisms of enhanced PD in the combination setting and enable translation of clinical dosing regimen.
Methods: Extensive single-agent and combination experiments to evaluate the PD effects of divarasib and migoprotafib were performed with H2122, a NSCLC cell line harboring the KRAS G12C mutation, both in an in vitro setting and in a mouse xenograft model. These studies quantitatively assessed the concentrations relationship with the following key PD endpoints: i) target occupancy by KRAS G12C alkylation, ii) mitogen-activated protein kinase (MAPK) pathway inhibition by the changes of phosphorylated extracellular signal-regulated kinase (pERK) in vitro, and the negative feedback regulation by transcript levels of dual-specificity phosphatase (DUSP6) and Sprouty (SPRY2) in mice tumor. A QSP model was adapted from literature [1] to include the following components: in vitro media free concentrations or mouse free drug pharmacokinetics, covalent binding of divarasib to KRAS G12C in its inactive state, reversible binding of migoprotafib to SHP2 protein, and MAPK pathway signaling components including ERK, DUSP6 and SPRY2. The PD effect of drug combination was implicitly inferred by the interactions of these signaling molecules.
Results: A total of 31 QSP model parameters were calibrated to simultaneously fit for both the in vitro and mouse PD data. The model adequately described the in vitro shift in divarasib concentration - response curves by the combination of migoprotafib, quantifying the enhanced KRAS G12C alkylation and pERK inhibition. The model also quantified the mice PD data, including a deeper alkylation, DUSP, SPRY inhibition, longer duration of response and slower rebound to the baseline by combination, compared to single-agent of each drug. The model revealed that the mechanism of migoprotafib on enhancing divarasib PD is mainly through enhancing alkylation.
Conclusions: A translational QSP model was established to characterize the preclinical PD of the combination of divarasib and migoprotafib. The model can be readily adapted as a platform model to quantify the combination PD of divarasib / migoprotafib with other MAPK pathway inhibitors.
Citations: [1] Sayama H, et al., Virtual clinical trial simulations for a novel KRASG12C inhibitor (ASP2453) in non-small cell lung cancer. CPT-PSP Vol 10, 864-877, 2021